This replication package contains the following elements.

1. cleaned survey data

files:
 surveydata_clean_study1.csv
 surveydata_clean_study2.csv
 column_name_descriptions.csv 

These contain the cleaned survey data for Study 1 and Study 2, plus a description of the meanings of the columns present in them.

2. analysis scripts

files:
 discrimination_analysis_pooled.R
 discrimination_analysis_study1.R
 discrimination_analysis_study2.R
 functions.R
 
These contain the scripts for all analyses and figures in the paper's main text and appendices. They are organized into three scripts depending on whether the analysis is of Study 1, Study 2, or both studies pooled. These and all other scripts are meant to be run with the main replication_package folder, i.e. the folder they are in, as the working directory. The functions.R script contains supporting functions for creating figures.

3. further analyses for Appendix A (representativeness) and D (descriptives)

files:
 representativeness_study1.R
 representativeness_study2.R
 descriptives_pooled.R

These scripts contain extensive code used in two of the Appendices, which would have been cumbersome if included in the analysis scripts above. They are used to analyze the representativeness of both study samples, and to create tables containing descriptives for the outcomes and covariates.

4. output folders

folders:
 figures
 tables
 
These folders contain the figures and tables created by the analysis scripts.

5. folders with additional data

folders:
 comments
 DK Statistik
 
These folders contain additional data used by the analysis scripts. "comments" contains two files providing hand-coded ratings of whether a respondent's comment on the treatment/evidence they saw should be coded as having no meaningful content, and whether the comment contains pro-minority emotions. These exist for Study 2 respondents only. "DK Statistik" contains the population data that we compared our study demographics to for the representativeness analysis in Appendix A.
